8 or ey al 
ESSE ee eI SP SENN ers ne tay 
1896 | ‘ ROS4Z AMERICANA: 3 
sufficiently complete, and these species are scarcely able to 
cause confusion, at least to collectors and authors who will take 
the trouble to consult good descriptions. These species are: 
R. setigera Michx. (R. rubifolia R. Br.), R. Carolina L., R. humilis 
Marsh. (incl. R. parviflora Ehrh. and R. lucida Ehrh.), 2. nitida 
Willd., R. foliolosa Nutt., R. gymnocarpa Nutt., and R. minutifolta 
Engelm. 
But besides these species there are others less well known, 
which have frequently given rise to confusion. These latter will 
be considered especially in the notes sent to the BoTanicaL 
GAZETTE. 
The genus Rosa has had the singular fortune of having been 
studied more than any other genus, and of having had its species 
become more obscure and less recognizable as the work upon 
them has multiplied, so that today the study of the genus is 
dreaded by the great majority of botanists. On account of the 
chaotic state to which the genus has been reduced by species 
makers, some students have concluded that there are no estab- 
lished boundaries between the species, and that it is useless to 
seek for constant characters with which to separate them. For 
some years I have not ceased to protest against this idea, which 
is radically false, and contradicts well-observed facts. I will 
continue to affirm that the true species of the genus Kosa are 
clearly characterized, and as distinct from each other as those 
of any other genus. 
One can attribute the deplorable state of the genus for half 
a century to two principal causes, viz., the condition of the col- 
lected material, and the desire of numerous amateurs and florists 
to discover a great number of new species in a small territory. 
In most genera the species are represented in herbaria by indi- 
viduals more or less numerous all of which usually show the char- 
acters necessary for good specific determination, so that one may 
compare individuals with each other, may distinguish dwarf and 
giant variations, and may form some adequate conception of 
the possible modifications by organs due to lack or excess of 
vigor. 
